August 2017 Mr. Spock, Don Draper and The Undoing Project – Lessons for Investment Managers and Asset Allocators TINA BYLES WILLIAMS CIO & CEO but are susceptible to certain biases. They then spent 30 plus years researching these biases through both thoughtfulness and experimentation which ultimately led to the development of a new field called Behavioral Economics. Lewis chronicles their journey brilliantly by detailing the duo’s work, relationship and thinking. One interesting aspect of the story is that economists did not wel- Michael Lewis’s latest book, The Undoing Project, tells the story come the new school of thought with open arms. Rather, they of Daniel Khanneman and Amos Tversky, two Israeli psycholo- defended their rationality assumptions so that the foundation of gists who studied how people make decisions and in the pro- their models would not be undone. However, Khanneman and cess changed the way many economists think. As good deci- Tversky were persistent in their quest, designing experiments sion making is at the core of good investment management, we to test their theories. Their experiments were built around un- thought it useful to spend some time on the topic, and hopefully derstanding why people made specific decisions. They gave in the process, shed some light on how good investment man- people problems to solve or choices to make and found that agers make decisions. people more often did not make the optimal or rational choice. What Khaneman and Tversky’s work showed was that humans’ First some background. Prior to Khanneman and Tversky the decision making process was fallible. standard economic paradigm, often referred to as the Neo Classical Economic Model, was that people were rational and They took this fallibility a step further and isolated root causes of made their decisions based on their economic best interest. The various decisions. Others have built on their work and there are premise was that people possess the knowledge, skill and abil- now dozens of identified behavioral biases that the human brain ity to calculate the best outcome possible and make decisions uses when processing decisions. These discoveries transcend accordingly. Papers were written, books published, and theo- economics to all decision making and have been applied to im- ries developed using this basic premise as the foundation. prove medical diagnoses, hiring practices, and other disciplines. There may not be a better way to further explain this concept, and its flaws, than pointing to a pair of television characters, What Does That Mean To Us? Star Trek’s Mr. Spock and Mad Men’s Donald Draper. Spock, the world’s most famous Vulcan, was famous for his devotion to Our interest is in the ramifications for investment decision mak- logic and his uncanny computational ability. He was the quint- ing and more specifically, how investment managers and asset essential “rational man” able to reach the optimal solution to allocators can incorporate the knowledge of behavioral biases any business problem. This is where Draper comes into the into their manager selection and overall investment process. story. Draper was a caricature of an ad man in the 1950’s & Clearly one way of combating these behavioral biases is to have 1960’s. Slick, handsome and mysterious, he was the creative a fully systematic process; but many pure fundamental manag- director of a fictional Madison Avenue advertising firm and used ers have successfully created hard limits within their own inter- his cunning to develop ad campaigns for some of our favorite nal procedures to evade such biases while still allowing for the products. Clearly brilliant at getting people to buy products, he potential alpha of idiosyncratic expert-driven stock selection. created ads that often coaxed people into ignoring their rational With this in mind, we thought it would be useful to recap a few side. Whether his guile would have been effective on a Vulcan known behavioral biases and how they apply to both invest- is a topic for TV fans to debate, but his approach seemed to have ment decisions and the manager selection process. an uncanny resemblance to real-life ad campaigns. The Endowment Affect is the tendency of people to place too So did the advertising industry know something that economists much value on what they have. It is generally manifested by did not? Khanneman and Tversky thought so. But what was demanding much more to give up an object when compared to truly extraordinary was that as scientists they sought to prove what they would pay to acquire the same object. In the context it. They contended that people do not make decisions rationally, of investing, a manager might overvalue the stocks that they Philadelphia | Chicago | San Francisco FIS GROUP | www.fisgroup.com | 215.567.1100 2 own in comparison to other alternatives. Many fundamental ing. One way to compensate is to ask all investment managers managers combat this bias by setting hard target prices for their under consideration to complete a questionnaire in the same stocks, while some go a step further by formally reevaluating format. As mentioned previously, understanding the impact of their holdings on a periodic basis to assess whether the invest- style effects vs. manager skill is also critical here. This allows ment thesis still holds. Some managers even rotate analyst evaluators to view all data the same way and removes some of coverage of specific names to ensure that a company is always the salesmanship from the selection process. being looked at with a fresh perspective. In the manager selec- tion realm, studying a manager’s “sell discipline” to understand Confirmation Bias is the tendency to weigh information that what triggers the sale of a position can help allocators under- supports an existing opinion more than information that does stand if managers compensate for this bias. Similarly, by es- not. In the manager selection world, relying on a first impres- tablishing preset procedures and evaluation criteria, allocators sion of an investment manager when subsequent data does can compensate for this bias that would otherwise lead them to not support that impression would be an example of confir- hold an underperforming manager too long. mation bias. Another (unfortunate) example is that large firms represented by portfolio managers and investment executives The Availability Bias states that people give too much weight that are cardboard cutouts of Don Draper are more consistent to items that are fresh (or available) in their memories. Ways to with the investment industry’s ethnic and gender expectations combat this bias are to assign equal weights to long-term and of competence than a small firm run or populated by the likes short-term factors in evaluation models, or to have disciplined of Claire Huxtable (the iconic professional working mom who buy and sell rules that clearly define preferences for stocks with co-starred in The Cosby Show). Because Claire represents an similar risk/reward ratios based on non-subjective criteria. In anomaly and Don does not; Don would likely receive more of the manager selection realm, putting too much emphasis on the “the benefit of the doubt” for a new strategy or a challeng- recent performance is the obvious example of this. To compen- ing performance period than Claire. Again, building an invest- sate, it is important to study a manager’s full track record over ment process that evaluates all aspects of a manager can help at least a five year period and to dissect what can be attributed compensate for this bias. to the style effects versus skill or its absence thereof. As Jer- emy Grantham once famously said, “90% of what passes for Conclusion brilliance or incompetence in investing is the ebb and flow of investment style”. While all of this may seem like common sense, it is not that simple. Prior to Khanneman’s and Tverskey’s work, no one had Framing is the tendency to draw different conclusions from identified behavioral biases. It was not necessarily an achieve- the same information depending on how it is presented. Some ment that they found that people make mistakes, but their con- managers adjust for this bias by expanding their research be- tribution was to understand the behavioral underpinnings of the yond corporate management communications to other firms mistakes. The good news is that this allows for course correc- within a company’s supply chain. Others have multiple analysts tion. or PMs meet with management and write-up their conclusions independently in order to provide more objectivity in the inter- So while all investment managers are susceptible to behav- pretation of the information provided. On the manager selec- ioral biases, the good ones can still employ a disciplined de- tion front, anyone who has ever been pitched a strategy from cision making process that will compensate for many of the an investment manager knows how good they can be at sales- biases. By applying the same persistence and thoughtfulness manship. Some managers’ ability to present themselves in a fa- as Khanneman and Tverskey, good investment managers and vorable light would even make Donald Draper proud. Isolating asset allocators can continually identify areas to improve their good performance periods, using inappropriate benchmarks investment process. That will ultimately lead to better decisions. and showing only profitable trades can be examples of fram- And as we said at the outset, good decisions are at the core of good investment management. Important Disclosures: This report is neither an offer to sell nor a solicitation to invest in any product offered by FIS Group, Inc. and should not be considered as investment advice. This report was prepared for clients and prospective clients of FIS Group and is intended to be used solely by such clients and prospects for educational and illustrative purposes.